Scientific research is often conducted in a highly unscientific manner. About $200 billion — or about 85 percent of global spending on research — is routinely wasted on poorly designed and redundant studies. As much as 30 percent of the most influential original medical research papers later turn out to be wrong or exaggerated.
But finally, there's a massive push to fix these problems — and it's largely being financed by a billionaire couple from Houston.
The Laura and John Arnold Foundation helped establish Metrics, a Stanford institute focused on "meta research" — or research on research — to identify problems in the scientific process. They've backed large-scale "reproducibility projects" dedicated to rerunning hundreds of important experiments in everything from neuroscience to psychology, to figure out which ones are actually reliable. They've even funded crusading scientists who want to make the research process more transparent and hold their peers to higher standards.
The Arnolds made a fortune (estimated at $4 billion) in finance in their 30s, and promptly started to give it all away, establishing their foundation in 2008. They aren't concerned with short-term rewards, like seeing their names on the side of buildings. Instead, they want to invest in solving long-term, systemic problems, including those in science. And they're filling a very important gap that's been left by traditional science-funding agencies, which often focus on innovation and breakthroughs.
I spoke to Laura Arnold, a former corporate lawyer and businesswoman, about her vision and why she became interested in fixing science.
Julia Belluz: In a short time, you’ve become the new patrons of research integrity, like the Medicis for nerds keen on improving science.
Laura Arnold: Most, if not all, of what we do is policy work of some form. We carefully think about market inefficiencies and big societal problems. We try to determine what the root causes are of those problems, and then we think of whether or not there are solutions that are feasible and logical for us as philanthropists to become involved in.
Then we think about how we can contribute to that solution by creating clear alternatives or clear contributions that would address that problem in a systemic way that’s sustainable.
As we started working in various fields, it became very clear to us that the state of research is not where it should be. All of us rely on researchers to tell us what’s healthy, to tell us what we should eat or not, what causes cancer or disease, to give us insights into what kinds of behaviors might lead to others, what attitudes might have certain effects.
Research affects everything we do, from medicine and pharmaceuticals to basic science. So if we don’t have a good core of rigorous research, and as a research community have discipline as to what constitutes rigorous research, almost by definition our work will not be accomplished. We will not be able to move the conversation toward logical good answers.
JB: One of the areas you support is meta research, or research on research. It's often overlooked by funding agencies because it's by definition pretty unsexy. Yet we know it's hugely important — to bring together all the knowledge we already have and figure out how reliable it is. How did you become interested in this area?
LA: Your average American will pick up the newspaper and see a headline that Tamiflu cures the flu or [coffee is good for your health]. A lot of these sorts of incredibly flawed correlational or observational studies are being touted as causation studies. You believe them.
We fundamentally believe that’s a mistake. So we started really becoming interested in this issue and thinking about how we could transform the space. We were fortunate enough to find Brian Nosek and the Center for Open Science. We become involved in reproducibility — to encourage the research community to do what science should do, not to be afraid to fail at experiments, to communicate those findings without fear of not catching a headline or not getting tenure.
We are trying to create an environment where researchers do the research in the most methodologically sound and honest way possible, to create an incentive structure or platform where the researcher could develop these ideas in a methodical and transparent way that benefits us all.
JB: It sounds like the Arnold Foundation is quite enlightened at a time when many other funders continue to emphasize innovation.
LA: I think the traditional philanthropy very often focuses on the breakthrough, things you can touch and feel the success: the scholarship, the building, the museum, the thing you can point to and say, "I contributed to that." Historically and traditionally, that has been the venue for much philanthropy. And there’s certainly a space and a need for that.
In our case, because of the combination of our backgrounds, the fact that we’re in fortunate a position to do this as a full-time job, that we were fortunate enough to do this at a young age, the opportunities we look at tend to be longer-term, transformative policy plays.
We’re less attracted to the breakthroughs, the most traditional analysis for this kind of investment. We’re much more attracted to what is a systemic problem that we can fix in the long term and will have a greater impact on this piece of research and all others.
JB: Some people have said you’re too unsentimental in your focus on data, that you seem to avoid feel-good projects. What do you say to that?
LA: We believe that policy should be informed by evidence. The reason we believe that isn’t because evidence is the end goal; it’s because we want to help people.
The reason we want to elevate the caliber of research — it’s not to criticize a certain researcher or see the standards for Nature change. The end goal is to help somebody with a disease make the right decision about a treatment, to give a doctor who is taking care of a child the ability and information to make the best treatment decision for that child, to help a person who is at risk of dying or of contracting diseases from obesity make the right lifestyle choices so he or she isn’t at risk.
So it is ultimately — and very prominently — about making society better and helping people.
We want social programs to help the highest number of people. We want governments to implement programs that work. How do you figure that out? You test and evaluate them. You put them through the method that yields a result that’s persuasive and conclusive.
So I’m not too bothered with the superficial comment about our data focus, because I understand and believe our data focus is precisely to help many more people than the few you could help with a direct grant.